| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Article |
Department of Medicine, University of North Carolina, Chapel Hill, North Carolina 27599
Abstract
IGF-I has been shown to enhance insulin sensitivity in patients with type I and type II diabetes. IGF-I suppresses GH, and this raises the question of whether its ability to enhance insulin sensitivity is mediated solely through a reduction in GH’s antiinsulin actions. This study was conducted to determine whether administration of a GH receptor antagonist to patients with acromegaly and insulin resistance would result in improvement in insulin sensitivity and whether IGF-I had any additional insulin-sensitizing effects over and above those induced by its ability to suppress GH secretion. Five patients with active acromegaly were treated for 2 wk with a GH receptor antagonist. The GH receptor antagonist was effective, as IGF-I fell 65%, and mean GH values rose 42%. Mean fasting insulin fell from 39 ± 6 to 30 ± 7 µU/ml, and this was accompanied by a 9% decrease in fasting glucose. After treatment the insulin sensitivity index was 2.7 ± 1.0 x 10-4/min·µU/ml compared with a baseline value of 1.65 ± 0.8 x 10-4/min·µU/ml (P < 0.015). Subsequently, the subjects were treated with the receptor antagonist plus IGF-I/IGF-binding protein-3 given by sc injection (1 mg/kg daily). After 2 wk of the combined treatment, fasting insulin fell from 49 ± 9 to 29 ± 7 µU/ml, and fasting glucose fell by 14%. The insulin sensitivity index improved to 4.34 ± 1.3 x 10-4/min·µU/ml, which was significantly greater than the value obtained after treatment with the GH antagonist alone. Although only a limited number of subjects were studied, the results strongly suggest that IGF-I has insulin-sensitizing actions that are independent of its ability to suppress GH secretion. These findings necessitate further studies into the non-GH-related mechanism by which IGF-I enhances insulin sensitivity.
ACROMEGALY IS CHARACTERIZED by GH hypersecretion and mild carbohydrate intolerance (1, 2). Although few acromegalic patients develop overt type II diabetes, many will develop elevated fasting insulin concentrations with normal or slightly increased fasting glucose values and impaired glucose tolerance (3, 4, 5). These abnormalities of insulin secretion and glucose dynamics often revert to normal after successful therapeutic intervention that returns GH secretion or action to normal (6, 7). It has generally been assumed that GH acting through its receptor leads to insulin antagonism (8, 9). This conclusion is based primarily on the results of animal studies, although administration of GH to GH-deficient children and adults clearly leads to blunted insulin sensitivity (10, 11). Whether these effects are actually mediated through the GH receptor (GHR) in acromegaly has not been definitively determined.
IGF-I is synthesized in response to GH administration, and its concentrations are elevated in acromegaly (12, 13). Administration of IGF-I has been shown in both experimental animals and humans to lead to enhanced insulin sensitivity (14, 15, 16). Recent gene-targeting studies in mice have shown that deleting hepatic IGF-I expression results in a 75% decrease in serum IGF-I, but it has no effect on the synthesis of IGF-I by peripheral tissues. This decrease in serum IGF-I is accompanied by increased insulin resistance, specifically in skeletal muscle (17, 18). Thus, serum IGF-I is believed to play a role in glucose homeostasis in experimental animals and humans by enhancing insulin sensitivity (19). Administration of IGF-I to subjects with impaired insulin sensitivity, primarily type II diabetics, has been shown to improve insulin sensitivity significantly (20, 21). IGF-I was shown in one study to enhance insulin sensitivity 3.4-fold, and in one trial with large numbers of subjects it was shown to improve hemoglobin A1c and lower insulin requirements in type II diabetics (20). Recently, these findings were confirmed using the combination of IGF-I plus its major serum binding protein (IGFBP-3) (22). When this combination was given to a group of type I diabetics, it resulted in a 54% reduction in mean insulin requirements and a 23% reduction in glucose. This treatment also resulted in a 4.8-fold reduction in nocturnal GH secretion. This raised the question of whether the IGF-I-induced reduction in GH secretion was the primary mechanism responsible for the improvement in insulin resistance. The recent development of a GHR antagonist has afforded us the opportunity to address this question (23). This study was undertaken to determine the role of blocking GHR activation with the subsequent reduction in serum IGF-I on insulin sensitivity in patients with acromegaly. This change in insulin sensitivity was then compared with the effect of the combination of the GH antagonist plus IGF-I/IGFBP-3.
Subjects and Methods
Patients
The study protocol was approved by the clinical research advisory committee and the committee on the protection of the rights of human subjects at University of North Carolina School of Medicine. Study patients were all adults previously diagnosed with acromegaly, and all gave informed consent. All subjects had failed conventional surgical and medical therapy and had persistently elevated serum IGF-I and GH concentrations. The patients were excluded if they had uncontrolled hypertension, proliferative diabetic retinopathy, or renal insufficiency, as determined by creatinine level greater than 2.0 mg/dl. Four of five patients had previously been treated with the GHR antagonist in an earlier study of the drug’s safety and efficacy (23). They had all been withdrawn from the drug for at least 6 months before the initiation of this study. None had been receiving any GHR antagonist therapy for 6 wk before the initiation of the study. The four previously studied subjects all had evidence of insulin resistance (elevated fasting insulin, >25 µU/ml) or overt type II diabetes (fasting glucose, >127 mg/dl). The fifth patient had a history of active acromegaly and had failed transsphenoidal surgery. She also had severe insulin resistance/type II diabetes. The patient characteristics are shown in Table 1
.
|
Baseline IGF-I, GH, insulin, and glucose levels were obtained at the first visit after a 14-h fast. Patients were then given their previous GHR antagonist dose (either 10 or 20 mg, sc, daily). The patient who was GHR antagonist naive was given a 20 mg/d dose. Patients were treated with the GHR antagonist for 2 wk. During that time they continued their usual therapy for diabetes and attempted to maintain their usual glucose control. After 2 wk they fasted overnight, then the next morning they underwent a frequently sampled iv glucose tolerance test to determine their insulin sensitivity index. Fasting IGF-I levels were obtained to document the effect of the GHR antagonist. Fasting blood samples for GH were collected 15, 10, 5, and 1 min before glucose administration.
The GHR antagonist was then discontinued for 2 wk, and subjects continued their usual diet and diabetes treatments. After that interval they were fasted overnight, then underwent a second iv glucose tolerance test to determine insulin sensitivity while off medication. Fasting IGF-I and repeated sampling for GH were repeated as described previously. The subjects then received injections of the GHR antagonist using the same dose that they had received previously. They were also given IGF-I/IGFBP-3 (1.0 mg/kg, sc, daily). After 2 wk they returned after an overnight fast, and the testing protocol was repeated as previously described.
Insulin sensitivity was assessed by the frequently sampled iv glucose tolerance test with minimum model analysis. Thirty-three plasma samples were obtained for glucose and insulin. Four samples were obtained before the administration of glucose, given in the form of a 50% solution (0.3 g/kg). At 20 min, 0.1 U/kg insulin was injected iv. After glucose administration, blood was collected at 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 20, 22, 23, 24, 25, 27, 30, 40, 50, 60, 70, 80, 90, 100, 120, 160, and 180 min for glucose and insulin measurements. The data were analyzed by minimum model analysis using the computer program MINMOD to determine the insulin sensitivity index. This technique has been used extensively to estimate insulin sensitivity (24, 25, 26).
Laboratory analyses
Glucose was measured by autoanalyzer (YSI, Inc., Yellow Springs, OH). Insulin values were determined by a two-site immunoradiometric assay using the manufacturer’s recommendation (Linco Research, Inc., St. Charles, MO). The lower limit of sensitivity of this assay is 0.2 µU/ml. Total serum IGF-I values were determined by double antibody RIA after extraction of IGFBPs by Sep-Pak (Millipore Corp., Bedford, MA) chromatography as previously described (27). This method results in the removal of all of the IGFBPs, but is associated with a 23% loss of IGF-I. Therefore, an IGF-I standard was run in parallel with each group of samples and used to correct for recovery. GH was measured by a sensitive chemiluminescence assay from Quest Diagnostics, Inc. (San Juan Capistrano, CA).
Statistics
All data are expressed as the mean ± SD. The significance of the effects of IGF-I/IGFBP-3 on various outcome parameters was determined using repeated measures ANOVA. This was followed by an unpaired Bonferroni test, which was used to make individual comparisons using the SAS program (SAS Institute, Inc., Cary, NC). Significant changes are expressed as P < 0.015.
Results
Administration of the GHR antagonist for 2 wk was accompanied by evidence of reduced GH target actions. Specifically, IGF-I fell from a pretreatment mean of 552 ± 194 to 309 ± 107 ng/ml (P < 0.01; Table 2). This decrease in IGF-I was accompanied by a modest increase in mean GH concentrations from 4.88 ± 1.2 to 6.62 ± 1.6 ng/ml, but this change was not significant. Thus, all five subjects exhibited a drug effect and had changes in serum IGF-I that are typical for blockade of the GHR activity using this antagonist (23). Changes in carbohydrate metabolism were also detected. Specifically, fasting insulin levels fell from 39 ± 6 to 30 ± 7 µU/ml (P < 0.01), and fasting glucose decreased 119 ± 11 to 109 ± 10 mg/dl (P = 0.08; Table 2). IGFBP-1, an index of insulin sensitivity, rose from 26.4 ± 5.8 to 47.9 ± 6.7 ng/ml (P < 0.015; Table 3). Most importantly, the insulin sensitivity index, calculated using the Bergman minimum model, showed a significant difference compared with this measurement after 2 wk without treatment (Fig. 1). At baseline this value was measured as 1.65 ± 0.8 x 10-4/min·µU/ml, whereas it was 2.7 ± 1.1 x 10-4/min·µU/ml, or a 64% increase, during GHR antagonist treatment. Weight remained constant during the GHR antagonist treatment interval and during the 2-wk period when the subjects received no medication.
|
|
|
). Despite the fact that IGF-I levels increased to values higher than those during the control period, GH levels showed no change (6.04 ± 1.4 compared with 5.15 ± 1.1 ng/ml; P = NS). This value was not different from that obtained during the control period, indicating that this incremental increase in IGF-I did not cause significant GH suppression. Parameters of insulin action also showed significant improvement during this phase of treatment. Specifically, fasting glucose fell from 118 ± 13 to 101 ± 9 mg/dl (P < 0.015 compared with control). Similarly, fasting insulin decreased from 49 ± 9 to 29 ± 7 µU/ml, and this decrease was significant (P < 0.015; Table 2). IGFBP-1 rose from 22.5 ± 5.7 to 53.9 ± 7.6 ng/ml (P < 0.01). Most importantly, the insulin sensitivity index increased to 4.34 ± 1.3 x 10-4/min·µU/ml (Fig. 1). This value was significantly increased over baseline (P < 0.01) and significantly greater than the value obtained with the GHR antagonist alone (P < 0.015). These changes indicate that IGF-I was able to further enhance insulin sensitivity beyond the effect noted with the GHR antagonist, and that the effect of IGF-I was not simply due to further suppression of GH. Weight was stable during this treatment interval, and there were no complications noted in the subjects who received IGF-I/IGFBP-3. Discussion
In normal subjects or patients with diabetes, GH is an important regulator of glucose metabolism and insulin sensitivity (28, 29, 30). Recent transgenic animal experiments in which GH is overexpressed have demonstrated that GH hypersecretion leads to increases in fasting glucose and insulin concentrations as well as decreased sensitivity to exogenous insulin administration. Both hepatic glucose output and peripheral glucose utilization are altered (31, 32, 33). Experiments in mice that have either deleted the GHR or overexpressed a GHR antagonist (the functional equivalent of a GHR knockout) have shown enhanced insulin sensitivity compared with normal mice (34) and resistance to the development of severe glucose intolerance when they are made diabetic (32). These findings suggest that GH binding to its receptor and subsequent GH-stimulated signal transduction lead to alterations in insulin sensitivity and, if GH hypersecretion is sufficiently severe, to an impairment in glucose utilization and insulin action. The molecular mechanism by which GH hypersecretion leads to impaired insulin action is unclear; however, one study showed that in GH-overexpressing animals, insulin receptor number in liver was reduced, and induction of receptor phosphorylation and phosphoinositol 3-kinase activation was attenuated in response to exogenous insulin administration (31). Therefore, enhanced GHR binding and GHR-linked signaling lead to alterations in peripheral and hepatic sensitivity to insulin. If these changes are sufficiently extensive, this results in impaired glucose metabolism.
The response to GH overexpression is difficult to interpret, however, because of concomitant increases in IGF-I. GH transgenic animals have increased serum IGF-I levels, and GH administration to GH-deficient children or adults leads to increases in IGF-I concentrations (10, 11, 30). Several studies have supported the concept that these increases in IGF-I that occur in acromegaly may counteract part of the antiinsulin actions of GH itself. IGF-I binding to its own receptor and binding to insulin/IGF hybrid receptors lead to enhanced insulin sensitivity (14, 20). The molecular mechanism underlying this enhancement is unclear; however, it has been clearly demonstrated in experimental animals and humans that administration of IGF-I to either normal subjects or patients with type I or type II diabetes leads to enhanced insulin action (14, 16, 20). One molecular mechanism that has been postulated to be important for IGF-I’s insulin-sensitizing actions is suppression of GH. Administration of IGF-I to normal subjects results in substantial GH suppression (35). However, when patients with GH resistance due to GHR mutations are given IGF-I, there is suppression of GH secretion, but there is also enhancement of insulin sensitivity (36), suggesting that IGF-I is exerting at least part of its effect by mechanisms other than by simply suppressing GH. In studies in type I diabetics, acute reduction in nocturnal GH secretion with pirenzepine was shown to result in improvement in insulin sensitivity (37). However, in type II diabetes this problem has not been analyzed. Most type II diabetics are overweight, and because of increases in adiposity and advanced age, they have minimal detectable GH secretion (38). One study did demonstrate significantly higher nocturnal GH secretion in type II diabetics with retinopathy, but this change did not correlate with the degree of insulin resistance in these subjects (39).
To eliminate the effect of the confounding variable of IGF-I-induced GH suppression, we used the recently developed GHR antagonist. Patients with acromegaly were chosen because they are likely to be the most extreme example of insulin resistance induced by GH. Patients with acromegaly have been shown to have elevated fasting insulin concentrations and insulin resistance, although there is a high degree of variability in this response (4, 5, 40). Additionally, many acromegalics have mildly impaired peripheral glucose metabolism, although hepatic glucose output is often normal (2, 40). When insulin is administered to acromegalics, the response is impaired in terms of both improving glucose clearance in muscle and suppressing hepatic glucose output (3, 40, 41). These findings suggest that many patients with acromegaly, although producing adequate insulin to compensate for the insulin resistance induced by GH hypersecretion, still have clear-cut manifestations of insulin resistance that are believed to be mediated through the GHR (42). This conclusion is further supported by the observation that successful cure of acromegaly with return to normal GH dynamics has been shown to alleviate the defects in insulin action (6).
Our findings clearly demonstrate that the GHR antagonist causes significant improvement in the insulin sensitivity index as well as fasting insulin and glucose concentrations. Although 2 wk may not have been adequate to achieve maximum improvement in insulin sensitivity, a significant change was noted. The subjects weight did not change; therefore, it is unlikely that major changes in body composition occurred that could have confounded the interpretation of changes in insulin sensitivity.
The addition of IGF-I in combination with the GHR antagonist resulted in substantially greater improvement in the insulin sensitivity index. The incremental change when the effect of administering the GHR antagonist alone was compared with that of administering the antagonist plus IGF-I was statistically significant. It should be noted that IGF-I increased to significantly greater values after administration of the GHR antagonist plus IGF-I compared with those during the initial control period. Therefore, we cannot exclude the possibility that these high levels of IGF-I were exerting an effect on insulin sensitivity that was greater than the effect of IGF-I in the basal untreated state. However, as the GHR antagonist induced a blockade of GH action that was adequate to normalize IGF-I, this suggests that IGF-I was exerting additional effects on insulin sensitivity that are not mediated through suppression of GH secretion. This was further confirmed by measurement of GH at the end of the treatment period, which showed that there was no significant difference between GH concentrations during treatment with the GHR antagonist alone compared with those after the combination of the receptor antagonist plus IGF-I. As GH hypersecretion was still present during both treatment intervals, it is clear that the additional increment in plasma IGF-I was insufficient to suppress GH into the normal physiological range. Thus, further suppression of GH was very unlikely to have accounted for the enhanced improvement in insulin sensitivity. Therefore, it is probable that under usual conditions the elevated IGF-I concentrations that occur in acromegaly are partially counteracting the insulin-antagonizing effects of elevated GH concentrations.
The molecular mechanism by which IGF-I is improving insulin sensitivity in addition to GH suppression is unknown. The theories that have been proposed include suppression of hepatic glucose output through hybrid insulin/IGF-I receptors (20) and enhanced insulin action in skeletal muscle, possibly due to hybrid receptor activation in that tissue (43). Federici et al. (44) have shown that hybrid receptors are markedly overexpressed in skeletal muscle in patients with type II diabetes, thus forming an excellent target for IGF-I action. As hybrid receptors contain an insulin receptor ß-subunit that is transphosphorylated by the IGF-I receptor ß-subunit, it is possible that activation of this receptor subunit is sufficient to activate normal insulin signaling in skeletal muscle and thus augment the ability of ambient insulin to stimulate glucose transport (45). In contrast, hybrid receptors have very low affinity for insulin, and even the higher insulin concentrations that occur in insulin-resistant states are unlikely to activate hybrid receptors in a manner comparable to the effect of pharmacological administration of IGF-I.
In summary, administration of IGF-I to patients with acromegaly results in improvement in insulin sensitivity in part by suppressing GH secretion and action, but also by enhancing insulin sensitivity independently of GH suppression. Future studies need to address the possibility that this phenomenon occurs in normal man and in patients with diabetes who have minimal or no GH hypersecretion.
Acknowledgments
We thank Laura Lindsey for her help in preparing the manuscript. We thank Pharmacia Inc. for providing the Somavert GHR antagonist. We thank Insmed, Inc., for providing SomatoKine IGF-I/IGFBP-3.
Footnotes
Address all correspondence and requests for reprints to: Dr. David R. Clemmons, 6111 Thurston Bowles Building, Department of Medicine, University of North Carolina, Chapel Hill, North Carolina 27599.
This work was supported by NIH Grants AG-02331 and RR-00046 and a grant from the Endocrine Fellows Foundation.
Abbreviation: GHR, GH receptor.
Received March 5, 2002.
Accepted July 3, 2002.
References
subunit ligand occupancy and ß subunit autophosphorylation in insulin/insulin-like growth factor-I hybrid receptors. J Biol Chem 268:7393–7400This article has been cited by other articles:
 |
|
 |
|
  N. Moller and J. O. L. Jorgensen Effects of Growth Hormone on Glucose, Lipid, and Protein Metabolism in Human Subjects Endocr. Rev., April 1, 2009; 30(2): 152 - 177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. Muzumdar, X. Ma, S. Fishman, X. Yang, G. Atzmon, P. Vuguin, F. H. Einstein, D. Hwang, P. Cohen, and N. Barzilai Central and Opposing Effects of IGF-I and IGF-Binding Protein-3 on Systemic Insulin Action. Diabetes, October 1, 2006; 55(10): 2788 - 2796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Rico-Bautista, C. J. Greenhalgh, P. Tollet-Egnell, D. J. Hilton, W. S. Alexander, G. Norstedt, and A. Flores-Morales Suppressor of Cytokine Signaling-2 Deficiency Induces Molecular and Metabolic Changes that Partially Overlap with Growth Hormone-Dependent Effects Mol. Endocrinol., March 1, 2005; 19(3): 781 - 793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Saukkonen, R. Amin, R. M. Williams, C. Fox, K. C. Yuen, M. A. White, A. M. Umpleby, C. L. Acerini, and D. B. Dunger Dose-Dependent Effects of Recombinant Human Insulin-Like Growth Factor (IGF)-I/IGF Binding Protein-3 Complex on Overnight Growth Hormone Secretion and Insulin Sensitivity in Type 1 Diabetes J. Clin. Endocrinol. Metab., September 1, 2004; 89(9): 4634 - 4641. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Haluzik, S. Yakar, O. Gavrilova, J. Setser, Y. Boisclair, and D. LeRoith Insulin Resistance in the Liver-Specific IGF-1 Gene-Deleted Mouse Is Abrogated by Deletion of the Acid-Labile Subunit of the IGF-Binding Protein-3 Complex: Relative Roles of Growth Hormone and IGF-1 in Insulin Resistance Diabetes, October 1, 2003; 52(10): 2483 - 2489. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Livingstone and G. A. Ferns Review: Insulin-like growth factor-related proteins and diabetic complications The British Journal of Diabetes & Vascular Disease, September 1, 2003; 3(5): 326 - 331. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
